Teaching hospitals are responsible for much of the clinical training for physicians before they enter independent medical practice. They tend to be large, usually urban, institutions and have a reputation for being expensive. Do they provide better quality of care? Research in the Journal of the American Medical Association suggests they do, at least by mortality measures. (JAMA Article) The researchers examined over 20 million Medicare admissions from 2012 to 2104 at 4483 hospitals, of which 250 were major teaching institutions and 894 were minor teaching ones. In unadjusted analyses of 30-day mortality rates, major teaching hospitals were at 8.1%, minor teaching at 9.2% and non-teaching at 9.6%. After adjusting for patient characteristics the gap grew slightly to 1.7%. After adjusting for hospital characteristics, it shrunk to 1.2%. Looking at the 15 most common reasons for medical admission, the gap in mortality between major teaching and non-teaching facilities was .7%. For surgical admissions, the difference was 1.2%. So teaching status is correlated with lower mortality for patients. The size of a hospital appears to have an impact as well, as large teaching hospitals had lower mortality rates than medium-sized ones, but large and medium-sized non-teaching institutions had identical mortality rates. Seven-day mortality was also slightly lower at teaching facilities, as was 90-day mortality.
The results lead to a well-powered conclusion that a Medicare patient is less likely to die following admission to a major teaching hospital. This is a fairly strong indicator of better quality. Why is less clear. The results did not appear to be volume-dependent. More use of advanced technology also seemed unlikely to be a reason for the difference. Better adherence to process of care measures may have been a partial explanation. As health plans often introduce narrow networks now, many are leaving out teaching hospitals due to cost. These networks may not offer patients the best possible outcome opportunity.